Professional Documents
Culture Documents
non prot neighborhood housing groups engaged in community development and organizing throughout the City. ANHDs mission is to ensure ourishing neighborhoods and decent, aordable housing for all New Yorkers. Leaders of the Citys community development corpora ons founded ANHD in 1974 to provide a unied voice for grassroots housing groups that focus on the needs of working class and low income neighborhoods. Over the past 39 years our membership has grown from eight founding members to todays 94 groups.
Contents
Acknowledgments...................................................................................................1 Execu ve Summary..................................................................................................2 Introduc on.............................................................................................................7 The New Housing Marketplace Strengths and Weaknesses...............................10 Where Has the Housing Been Built?.......................................................... 11 Housing for What Size Family?.................................................................. 15 Aordable to Whom?................................................................................20 When Will the Aordability Be Lost?........................................................ 34 Beyond Housing How Can We Build Stronger Neighborhoods?........................................... 36 Looking Ahead:.......................................................................................... 39 The Real Aordability Index: A New Way of Measuring Return on Investment ......................................40 Appendices............................................................................................................ 46 Methodology............................................................................................. 46 New Housing Marketplace Units by Area Median Income (AMI), FY2004 FY2011..........................................................................................51 New Housing Marketplace Units by Community District, FY2004 FY2011.......................................................................................... 52 New Housing Marketplace Units by Size and Community District, FY2004 FY2010.......................................................................................... 54 New Housing Marketplace Units by Neighborhood Aordability, FY2009 FY2011..........................................................................................56 ANHD Member Organiza ons............................................................................... 57
Acknowledgements
ANHD would like to acknowledge the generous support of the following founda ons that made this report possible:
REAL AFFORDABILITY |
Execu ve Summary
T
Two thirds of New Housing Marketplace units are too expensive for the majority of local neighborhood residents.
he a ermath of Hurricane Sandy has underscored the challenges facing the leaders and residents of New York City. In the long term, we know that resilient neighborhoods need an ongoing commitment to strengthen both their physical and civic infrastructure so we are be er able to avoid damage and cope with the a ermath of future disasters. In the short term, thousands of people whose homes or businesses were destroyed or damaged will need to nd places to live and resources to rebuild their homes, businesses and lives. The community development community, in conjunc on with the City and the Federal Emergency Management Agency, has come together to nd and relocate all those in need as quickly as possible. However, in a city with rental vacancy rates as low as two percent, this is an enormous challenge. Many families need long term aordable housing solu ons, which were in short supply and very high demand prior to the storm. Hurricane Sandy has only highlighted the impor tance of ensuring robust aordable housing opportuni es for New York City residents. For all of our families an aordable safe home is the founda on to beginning to build up lives and move forward to a be er future.
Mayor Bloombergs New Housing Marketplace (NHMP) aordable housing development plan is an impressive achievement that took great strides towards crea ng aordable housing opportuni es for residents. It is on track, and will likely meet its goal of developing 165,000 units of aordable housing for New York City. NHMP now stands apart as the largest municipal aordable housing eort in the na ons history.1 The Mayors commitment and perseverance, even through a severe economic downturn, makes the achievement all the more remarkable. Yet despite the thousands of housing units created and the mul billion dollar investment, many City residents and housing advocates believe that the Bloomberg housing produc on plan was implemented with signicant aws that undermine the actual impact of the investment on local neighborhoods. The future leadership of our City must take into account the lessons of this experience in order to develop a more eec ve aordable housing produc on policy that brings greater impact on our local communi es and greater value for the taxpayer investment. The New Housing Marketplaces weaknesses are not in the number of units created they are in how well these units match the Real Aordability needs of New York City communi es. While the City has commi ed to and developed a signicant number of aordable housing units under the Bloomberg administra on, about two thirds of New Housing Marketplace units are too expensive for the majority of local neighborhood residents.
1 For the purposes of this report units built, created, or developed under NHMP refers to all NHMP units both new construc on and preserva on.
2 | REAL AFFORDABILITY
ANHD is commi ed to the preserva on and development of aordable housing in New York City. We work to ensure vibrant, aordable neighborhoods through advocacy for policies and programs that preserve and develop aordable housing for city residents. To prepare this report, ANHD completed an in depth evalua on of the Bloomberg administra ons New Housing Marketplace to understand the strengths and weaknesses of the ini a ve. This analysis looks beyond the singular units produced measure reported by the City. Our analysis looks citywide and community by community to examine unit size, length of aordability, depth of aordability, geographic loca on and community impact, all central components in whether a unit is truly accessible to a family or community. Using this comprehensive approach, ANHD was able to assess the Bloomberg housing programs success in providing Real Aordability opportuni es to its ci zens. ANHDs ndings include: NHMPs units too o en do not meet the actual aordability needs of the neighborhoods in which they were built. Approximately one third of NHMP units have an upper income limit above the actual New York City median income. And in half the Citys community districts, the majority of units built in the community are too expensive for a household earning the local median income for the neighborhood. City housing policy, including NHMP, has not preserved the longterm aordability of the units that are built. Star ng in 2017, the City will be at risk of losing an annual average of 11,000 units built with City subsidy and by 2037, the City could lose the aordability of as many units as were built by NHMP, greatly undermining the value of the Citys eorts. NHMP units were concentrated in a few community districts, although generally for valid development reasons. Five of the Citys 59 community districts account for 30% of all NHMP units built or preserved. The NHMP unit sizes largely do match the distribu on of household sizes for NYC. Two , three , and four person households make up 56% of New York City households, with 58% of the NHMP units appropriate to house these size households.
Depth of Aordability:
Length of Aordability:
Loca on:
Household Size:
REAL AFFORDABILITY |
The New Housing Marketplaces singular focus the 165,000 unit goal may have eclipsed the broader mission of crea ng aordable housing opportuni es that meet the needs of low and moderate income New York City residents. These Real Aordability factors must be to be taken into account in future city aordable housing policies and programs. Specically, ANHD is recommending:
1 2 3 4 5
The rent and income levels targeted by future aordable housing developments, while s ll serving a wide range of households, need to be be er targeted to community need right now, much of the housing built under the New Housing Marketplace is unaordable to the majority of households in the neighborhood in which it is built.
Future affordable housing should be affordable in perpetuity the City should not invest in short term aordability when it has the ability to require truly aordable housing for this and future genera ons for li le to no addi onal public subsidy. The City needs to provide aordable housing for every neighborhood while in many neighborhoods there remains less vacant City owned land that can be used for aordable housing development, there are a wide array of untapped resources and opportuni es for a new aordable housing development focus that can be explored in order to expand the geography of aordable housing development beyond only a few neighborhoods.
Not all aordable housing units are created equal in terms of community benet and taxpayer value.
The City needs to con nue to build a wide range of apartment sizes targe ng large and small families, as well as non family households and individuals living alone. The City needs to view housing development not just as providing units, but as comprehensive neighborhood and community development. Benets such as aordable retail op ons, community services, and environmentally friendly elements need to be part of our housing developments whenever possible. In order to incorporate this more robust Real Aordability approach, the City needs a new way of measuring success, and a new way of measuring return on investment. We know that not all aordable housing units are created equal in terms of community benet and taxpayer value. The next administra on must go beyond simply units pro duced as a measure of success, and look at the quality and type of these units. To meet this need, ANHD has developed the Real Aordability Index, a new way of measuring the impact of New York Citys aordable housing development and preser va on projects. By simply coun ng units, produc on incen ves are skewed toward crea ng smaller, less aordable units, which are supercially less expensive to create but do not necessarily provide the most public benet. This Index shi s the focus away from a uniform, singular unit count to a more complete model that takes into account a variety of factors that inuence the Real Aordability value of a development.
4 | REAL AFFORDABILITY
We hope the Real Aordability Index will help shape the future of City housing development policy and programs by encouraging the City not to simply develop the largest number of units, but to have the greatest eect on the community for each unit and leverage the highest return to the taxpayer. It incorporates a number of indices to present a more complete analysis of past and future developments aordability benet to the community and City. These indices are: Depth of Aordability: Length of Aordability: The income qualica ons and restric ons for households seek ing to reside in a development, as well as the actual rent or mortgage paid. The number of years a unit remains income restricted and accessible as aordable housing. The loca on of a development, looking at concentra on and distribu on of the areas aordable and market rate rental op ons, and the community need for greater aordability. The number of residents served by a development given the number of units and unit sizes. The value brought to residents and the neighborhood beyond housing, through addi onal commercial, community, or other mixed uses included as a part of the development.
Loca on:
Household Size:
Community Impact:
As an example, consider two very dierent aordable housing units, both developed under NHMP, both marketed as aordable housing at the same me (January 2011), and in the same neighborhood (Manha an Community District 10 Central Harlem). Unit A is a studio apartment, rents for $1,492 and the units aordability restric ons expire in 30 years. Unit B is a three bedroom apartment, rents for $531 and because its developed by a community based not for profit will remain affordable permanently, bene ng genera ons to come. While there may be residents that need both subsidized units, they address two very dierent households needs: one unit serves a single resident earning as much as $97,125; the other serves a low income family of four, ve, or six people making no more than $36,760 slightly more than one third as much as the maximum income for the single person. While both of these units represent an overall success of the New Housing Marketplace in a city like New York, every unit created is needed and does count clearly the second unit addresses a greater need and has a greater community impact. These types of dierences need to be counted when determining the amount of public resources that should be spent on dierent types of aordable housing.
REAL AFFORDABILITY |
Aordable housing policy should be wri en in a way that most eec vely and eciently responds to the diverse needs of NYCs varied communi es and residents.
This new metric is rooted in the belief that aordable housing policy should be wri en in a way that most eec vely and eciently responds to the diverse needs of NYCs varied communi es and residents. When calcula ng which developments are be er uses of public resources, considera ons like unit sizes, depth of aordability, geographic loca on, length of aordability, and other community benets that the development may bring need to be looked at, instead of just the total number of units produced. Addressing the Citys growing aordable housing crisis must be a comprehensive, nuanced, and three dimensional approach for the future. ANHD is commi ed to making a more robust understanding of aordable housing needs a central part of the conversa on, programs and policies going forward.
6 | REAL AFFORDABILITY
Introduc on
he impact of Hurricane Sandy has been extremely challenging for New Yorkers and has aected all of our lives. Local community groups organized an unprece dented response from Day 1, not only taking charge locally, but reaching across neighborhood and borough boundaries to deliver support to areas in need. Thousands of people whose homes were destroyed or made uninhabitable without basic services will need to nd new places to live for the coming weeks and months. The community development community, in conjunc on with the City and the Federal Emergency Management Agency, has come together to nd and relocate all those in need as quickly as possible. However, in a city with rental vacancy rates as low as two percent, this is an enormous challenge. Many families need long term aordable housing solu ons, which were in short supply and very high demand prior to the storm. Hurricane Sandy has only highlighted the importance of ensuring robust aordable housing opportuni es for New York City residents. For all of NYC families an aordable safe home is the founda on to beginning to build up their lives and move forward to a be er future. Mayor Bloombergs aordable housing development plan, the New Housing Marketplace, is an impressive achievement. It will most likely meet its goal of developing and preserving 165,000 aordable housing units for New York City. This mul billion dollar investment now stands apart as the largest municipal aordable housing eort in the na ons history. The Mayors commitment and perseverance, even through a severe economic downturn, makes the achievement all the more remarkable. Despite the thousands of housing units created and the billions of dollars invested, many City residents and housing advocates feel the Bloomberg housing produc on plan has been implemented with signicant aws that undermine the actual impact of the investment on local neighborhoods. New York Citys leadership must take into account the lessons learned from this experience in order to develop a more eec ve aordable housing produc on policy that can have greater impact on local communi es and greater value for the taxpayer investment. For too many New York City families and communi es, the Citys sizeable investment has not had the maximum impact on their community. Local housing advocates have frequently noted a stark mismatch between the units produced by the City and the needs of the residents in the communi es they serve. As one local not for prot aordable housing developer explained, For our last lo ery, we received thousands of applica ons from interested tenants. The vast majority of them, however, did not earn enough to meet the Citys guidelines. These concerns, voiced by both residents and housing advocates, capture the disconnect between the Citys momentous goal and the on the ground reality. While the New Housing Marketplace met an unprecedented benchmark, it misaimed by not taking
Hurricane Sandy has only highlighted the importance of ensuring robust aordable housing opportuni es for New York City residents.
REAL AFFORDABILITY |
into account the various aspects of aordable housing development that make a unit truly impac ul for a family and the community. This is largely due to the fact that the City uses a single, linear measure to gauge the impact and eec veness of their housing produc on program: the number of units created. Yet, we know that not all aordable housing units are created equal. A subsidized studio apartment ren ng for $1,492 that is only aordable for 30 years is dierent from a permanently aordable, three bedroom apartment ren ng for $531. The second unit: a) Houses more people a household of 4 to 6 persons versus a single individual. b) Is aordable for a longer me period permanent aordability versus expiring in 2041. c) Serves a household with a substan ally lower income and very limited op ons in the Citys expensive rental market one with a total annual income of no more than $36,760 as compared to subsidizing an individual earning as much as $97,125. But yet, both of these units were iden cally counted and celebrated by the City as one aordable unit successfully developed. While producing a signicant number of units is a very important part of addressing the Citys aordable housing shortage, the actual impact of each unit on the local community and the value created for the taxpayer also needs to be understood. This means including the depth of aordability, the length of aordability, the number of people served per household, and the addi onal services and benets the housing brings. In the future, the City must take a more nuanced, neighborhood specic approach that reects the type and quality of the housing, not just the amount built. The next mayor must tailor the aordable housing programs and policies to be er reect the Citys demographics and respond to the varied needs of residents and communi es across the ve boroughs. And the City must work to be er maximize the value of the aord able housing eort and investment to the taxpayer and the community. This paper presents the nding of ANHDs evalua on of the type of housing constructed and preserved during the New Housing Marketplace, based on newly compiled data, and sheds light on why many communi es believe the Mayors plan has not done enough to respond to the shortage of aordable housing most needed in their neighborhoods and the City as a whole. The analysis is broken into sec ons that inves gate where housing was built, the size of units, the depth of the aordability, and the length of aordability. Each sec on is followed by a series of recommenda ons that will help the next mayor be er ensure that aordable housing matches the public need. They outline alterna ves to the limi ng number of units produced measure of success, and introduce policies that, combined, present a comprehensive approach that generates Real Aordability for neighborhoods.
In the future, the City must take a more nuanced, neighborhood specic approach that reects the type and quality of the housing, not just the amount built.
8 | REAL AFFORDABILITY
Following the New Housing Marketplace analysis is a new ANHD developed index that can help assess the Real Aordability value of a housing development. This index assesses the public benet for each dollar of public investment, and measures the return to the taxpayer for dollars invested in aordable housing for past as well as proposed projects. This report will allow a broader discussion on how New York City both the current and future administra ons should align its resources and aordable housing policy in a way that most eec vely and eciently responds to the needs of our diverse communi es and residents. Going forward, we are commi ed to making a more comprehensive approach that includes unit size, length of aordability, depth of aordability, and community impact as central components in the Citys ability to provide Real Aordability opportuni es to its ci zens.
Going forward we are commi ed to making a more comprehensive approach that includes unit size, length of aordability, depth of aordability, and community impact as central components in the Citys ability to provide Real Aordability opportuni es to its ci zens.
REAL AFFORDABILITY |
ANHD would like to thank Meryl Block Weissman at HPD and Melissa Barkan at HDC for facilita ng these data requests.
10 | REAL AFFORDABILITY
Almost half of the community districts (28 of 59) have less than 1,000 units, leaving each of them with less than one percent of all units developed under the NHMP.
11
To get a be er sense of the NHMP development disparity from one community district to the next, ANHD took a closer look at one borough. Table 1 below shows all the NHMP housing developed in Brooklyn by the City of New York in each of its 18 community districts. Brooklyns 30,224 units of aordable housing represent about a quarter of the total units developed under the Mayors plan; the third highest borough a er Manha an and the Bronx. Table 1 clearly shows that the vast majority of Brooklyns NHMP units are concentrated in just a few community districts. Four of Brooklyns community districtsBedford Stuyvesant (BK3), Cypress Hills / East New York (BK5), Brighton Beach / Coney Island (BK13) and Brownsville / Ocean Hill (BK16) account for over 50% of the en re boroughs aordable housing development. Meanwhile, Bensonhurst (BK11), East Flatbush (BK 17), and Flatlands / Canarsie (BK 18) combined make up just over one percent of the boroughs units.
TABLE 1. NHMP
Community District
BK1 BK2 BK3 BK4 BK5 BK6 BK7 BK8 BK9 BK10 BK11 BK12 BK13 BK14 BK15 BK16 BK17 BK18 Total Greenpoint /Williamsburg Brooklyn Hts /Ft. Greene Bed Stuy Bushwick Cypress Hills/ENY/Starre Carroll Gardens/Gowanus/PS Sunset Park/ Windsor Terrace Crown Hts /Prospect Hts Crown Hts /Leerts Gardens Bay Ridge/Dyker Hts Bensonhurst /Gravesend Borough Park /Kensington Brighton Beach /Coney Is. Ditmas Park /Flatbush Sheepshead Bay /Gravesend Brownsville /Ocean Hill East Flatbush Flatlands /Canarsie
% Boro
9.6% 9.7% 10.5% 7.7% 17.4% 1.2% 1.5% 4.9% 4.6% 1.0% 0.1% 0.8% 10.3% 1.5% 4.1% 13.8% 0.8% 0.5% 100%
% City
2.3% 2.7% 2.5% 1.9% 4.2% 0.3% 0.4% 1.2% 1.1% 0.2% 0.1% 0.2% 2.5% 0.4% 1.0% 3.7% 0.2% 0.1% 25%
There are understandable explana ons for the geographic dispari es within boroughs and across the City. Certain constraints impact a community districts capacity for development, including land availability or zoning restric ons. City owned vacant land, a main source of real estate for aordable housing, has decreased rapidly over the last two decades. According to an analysis by ANHD, HPD owned vacant land can support approximately 7,537 more units of housing under current zoning. Of these 44% are in a
REAL AFFORDABILITY | 12
single Community District, Queens CB 14 (Rockaway Peninsula). Given Hurricane Sandys impact on the Rockaways the City will need to rethink if and how this HPD vacant land could be be used for new aordable housing. Another source of aordable housing development, city owned, tax foreclosed housing (in rem), has all but disappeared. As of June 2010, the Citys in rem stock was only 719 units.
FIGURE 2:
Data Source: NYC Department of Informa on Technology and Telecommunica ons, Open Data Project
The success of the NHMP was that it took full advantage of the development opportu ni es available, and the result was an unprecedented construc on and preserva on program leading to over a hundred thousand units of aordable housing. However, all neighborhoods need solu ons. Aordable housing is needed throughout the city, not just in certain areas. While some sources for aordable housing development are diminishing, the city can also access new opportuni es. New pipelines could enable aordable housing to be built throughout the city, developing equitable housing opportuni es that both solve a citywide problem and build poli cal support for aordable housing across the ve boroughs.
REAL AFFORDABILITY |
13
1 2
Aordable housing leveraged through the zoning code. The City con nues to see enormous development and growth from changes in zoning regula ons. This presents an extraordinary and untapped opportunity to systemically incorporate aordable housing into the citys growth. The Bloomberg administra on rezoned over 10,000 City blocks; aordable housing must be a part of the next 10,000 blocks that are rezoned. Policies such as inclusionary zoning have only scratched the surface of the possibili es. Through innova ve policies like mandatory inclusionary zoning, the City has the ability to create more aordable housing each me a neighborhood is rezoned at no addi onal cost to the City. These alterna ve sources and pipelines have the advantage of not necessarily being narrowly geographically focused, as is vacant HPD owned land. This allows the next administra on to develop a truly comprehensive housing plan that includes strategies specic to each of the ve boroughs. The reality is that in New York City, aordable housing is needed everywhere; in all of our neighborhoods, rents and housing prices are increasingly unaordable for the local popula on. The next administra on needs to recognize, and respond, to the housing crisis in all of our communi es.
14 | REAL AFFORDABILITY
The New Housing Marketplace has mostly met the needs of New Yorkers in the size of the apartments it has built.
1 BR 31.0%
One of New York City residents top cri cisms of NHMP has been that the units developed are not large enough for their families. This cri cism depends heavily on a given familys size. For families of three and some mes four, a two bedroom apartment is sucient. However, all families of ve or more, some families of four, and even some families of three, need a three bedroom apartment or larger. Therefore units that are two bedrooms or larger are o en dis nguished as family size from SROs, studios or one bedrooms, which serve individuals or couples. By grouping the data from Table 2, ANHD found that about half (47%) the NHMP units serve individuals and couples (i.e., SRO, Studio, or 1 bedroom apartments) and about half (50%) the units are family size. (i.e., 2+ bedrooms). The remaining 3% of the units were of an unknown size.
3 The sample includes the following units: all HPD Preserva on and HPD New Construc on projects started prior to July 2010. A group of pre July 2010 New Construc on projects are not in this le if they were completed in the past two years. 4
ANHD paid HPD to re congure its database to obtain some of this data.
REAL AFFORDABILITY |
15
ANHD then compared the FY2004 FY2010 set of New Housing Marketplace unit sizes to the household sizes of New York City, in order to determine if the unit sizes were in alignment with the popula on at large. To do so, ANHD applied a standard of 1.5 people per bedroom, with 1 person for a studio or SRO. Table 3 explains how this standard was applied to unit sizes in order to generate the number of persons served at each unit size.
TABLE 3: ANHD Calculated Household Size by Unit Size
ANHD Calculated Household Size 100% to 1 person households. 100% to 1 person households. 50% to 1 person households, 50% to 2 person households. 25% to 2 person households, 50% to 3 person households, 25% to 4 person households. 16.7% to 3 person households, 33.3% to 4 person households, 33.3% to 5 person households, 16.7% to 6 person households. 25% to 6 person households, 75% to 7+ person households. 100% to 7+ person households. 100% to 7+ person households.
Despite the communitys cri cisms around a lack of family sized units, ANHD found that the size of units constructed in the FY2004 FY2010 New Housing Marketplace sample largely mirror New York household sizes. As Figure 3 shows, there was a slight undersupply of units for two person households, and a slight oversupply of units for three person households. But the pa ern of the New Housing Marketplace is clear; its apartment size distribu on closely reects the household sizes of New York City a prac ce that should con nue. However in the future, City data should include unit sizes so that a more detailed analysis can be done. For example, the limited data does not allow ANHD to separate unit size by preserva on vs. new construc on, or by household income, or rent/own status. This greatly limits ANHD and other housing advocates ability to understand how unit size inuences accessibility to aordable housing for New York City communi es and families.
16 | REAL AFFORDABILITY
As previously stated, a neighborhood perspec ve must be taken into account in each analysis. Looking across boroughs, the distribu on is less uniform. For example, Manha an has a larger share of smaller units (54%) compared to family size (45%). Queens, on the other hand, has a much larger share of family size housing (71%) compared to smaller units (29%). When examining bedroom size by community district, the distribu on becomes even less uniform. Table 4 (below) shows the unit size distribu on for the 14 Community Districts of Queens. We can see that the unit size distribu on varies greatly by Community District, with East/Central Queens (QN 8) featuring 100% small units, and Ridgewood/Maspeth (QN 5) featuring 100% family units. Narrowing the focus to community districts with at least 100 units developed, the range is from 86% small units in Astoria (QN 1), to 89% family units on the Rockaway Peninsula (QN 14), with the remaining community district, Jamaica/St. Albans (QN 12), about evenly split, at 46% small units and 54% family units.
TABLE 4: Number of Bedrooms by Community District in Queens
Studio
QN 1 QN 2 QN 3 QN 4 QN 5 QN 6 QN 7 QN 8 QN 9 QN 10 QN 11 QN 12 QN 13 QN 14 TOTAL 32
1 BR
230 16 3 1
2 BR
39 9 8 2 4
3 BR 4 BR
2 1 1 1 6 2 8 87 7 762 877
5 BR
Total
303 17 12 10 3 10 81 17 14 319 14 1,635
0 1 BR
262 16 3 1
2+ BR
41 1 9 9 3 10
1 4 78 5 696 846
2 1 8 2 11 2
2,435
17
In addi on to the FY2004 FY2010 unit size dataset, ANHD was able to gain access to a separate dataset of FY2010 and FY2011 New Housing Marketplace projects that included more detailed informa on about each unit. This FY2010 FY2011 data includes 9,970 units that were not a part of the 53,211 units in the FY2004 FY2010 sample discussed above. It allowed us to analyze both bedroom size and AMI levels.6 Table 5 (below) summarizes this data by number of units. The 51% 60% AMI band was well represented irrespec ve of unit size, with studios, one bedroom, two bedroom, and three bedroom apartments all in the top 10 highest subgroups.
TABLE 5: Number
AMI
0% 30% 31% 40% 41% 50% 51% 60% 61% 65% 66% 70% 71% 80% 81% 90% 91% 100% 101% 110% 111% 120% 121% 130% 131% 140% 141% 150% 151% 160% 161% 170% 171% 180% Non restricted Super
1 BR
65 163 320 1,606 2 9 401 26 104 18 10 146 43 58 345 2 3,318
2 BR
41 150 179 2,009 11 4 583 74 165 83 6 180 65 88 1 173 65 3,877
3 BR
24 22 46 294 6 93 174 104 28 2 75 22 6 30 16 54 4 1,000
4 BR Unknown Total
2 174 457 848 4,514 14 19 1,245 298 395 134 18 445 216 6 0 282 17 816 72 9,970
Total
18 | REAL AFFORDABILITY
Data is from a HPD applica on called OnTrack, a web based tracking applica on used by HPDs Oce of Development to track the loan closing process. The data in these tables was collected by OnTrack for New Construc on projects started a er July 2010 and a group of pre July 2010 New Construc on starts if they completed in the past 2 years.
REAL AFFORDABILITY |
19
Aordable to Whom?
The Depth of Aordability, or the household income necessary to qualify for an aordable housing unit, is perhaps the most striking shortcoming of current aordable housing development in the City, where units produced is the only measurement of success. New Housing Marketplace makes no dis nc on between the value of a unit targe ng a family of four earning a household income of $24,900 (30% of HUD determined Area Median Income) versus a unit targe ng a family of four earning a household income of $136,950 (165% of HUD determined Area Median Income). Under the current measure of a per unit count, these two apartments developed under NHMP are equally applauded. A result of this policy is that much of the New Housing Marketplace has been geared toward the wealthier half of New York City.
Due to the at units produced measure, the City has not always categorized aordable housing by specic income level. Instead, units are labeled as follows: (1) Low income (<80% of AMI) (2) Moderate income (80% 120% of AMI) (3) Middle income (120% 180% of AMI).7 Since a large percentage of units developed by the City are nanced with Low Income Housing Tax Credits (LIHTC), the City also tracks the less than 60% AMI category. Based on these AMI categories the NHMP can report successfully building and preserving over 100,000 units for low income residents (see Table 6). This is because over a third of the units (34%) developed under the NHMP are targeted to households earning less than 60% of AMI. And 4 out of 5 units developed by the NHMP (83%) are targeted to households earning less than 80% of AMI.
TABLE 6: All NHMP Units by AMI, FY2004FY2011
0 60% 61 80% MODERATE 81 100% 101 120% MIDDLE 121 180% Non Restricted Super Other Total
LOW
NHMP Number Percent 42,513 34.2% 60,580 48.7% 6,356 5.1% 1,842 1.5% 10,431 8.4% 2,273 1.8% 182 0.1% 241 0.2% 124,418 100%
20 | REAL AFFORDABILITY
7 HUD rental housing programs generally dene low income as less than 80% of their adjusted AMI levels, and moderate income as less than 120% of their adjusted AMI levels, and this is the terminology the New Housing Marketplace uses as well. Other programs use dierent deni ons. For instance, when determining low and moderate income census tracts for Community Reinvestment Act purposes, low income is dened as less than 50% of AMI. Moderate income is dened as 50% 80% of AMI (which the New Housing Marketplace calls low income), and 80% 120% of AMI is dened as middle income (which the New Housing Marketplace calls moderate income). Some programs dont use a percent of AMI at all, and instead use dierent indicators, such as mul pliers of the poverty level (usually 150%). The HUD rental housing, and as such the New Housing Marketplace, deni on of low , moderate , and middle income is higher than any other governmental deni on of the terms.
In FY2009, the City began tracking household income levels in 10% increments, which allows for a much more robust analysis of aordability categories. Table 7 details those units serving households earning less than 80% AMI developed under NHMP between FY2009 and FY2011. Instead of NHMPs broad low category, we are able to break down aordability into below 30%, and 10% increments between 30% and 80% AMI. Looking at Table 7, we see that in the past three years of NHMP, the City developed just 3,049 units (7.9% of all FY2009 FY2011 units) for households earning less than 40% AMI, even though this income cohort represents nearly a third of all New York City households. At the same me, the NHMP developed 21,507 units TABLE 7: Low Income NHMP Units by AMI, (56% of all FY2009 FY2011 units) FY2009FY2011 for those making 51% 80% of AMI, NHMP an income cohort that represents Number Percent just 17% of New York City house 0 30% 1,389 3.6% holds. So while the NHMP did 31 40% 1,660 4.3% primarily serve households below 41 50% 4,606 11.9% 80% AMI, most of these units were targeted at those at the upper end 51 60% 14,907 38.6% of this spectrum, and a small share 61 70% 2,895 7.5% of units were targeted at those 71 80% 3,705 9.6% households most in need. Table 8 (below) shows the number and percent of below 80% AMI units for Community Districts in Manha an. Eight of the twelve Community Districts in Manha an did not see any housing developed for households in the lowest income cohort, those earning less than 30% AMI. The four that did Chinatown/ Lower East Side (MN3), Hamilton Heights / West Harlem (MN9), Central Harlem (MN10), and East Harlem (MN11) are among the poorest neighborhoods in the borough. In another four Community Districts, Financial District (MN1), West Village / SoHo (MN2), Union Square (MN5), and Stuyvesant Town / Turtle Bay (MN6), no units aordable to households earning less than 50% AMI were built. And in two of these areas (MN1 and MN6), two of the boroughs most auent neighborhoods, there was not a single unit built for any resident making less than 80% of AMI.
TABLE 8:
Number of NHMP Low Income Units in Manha an by AMI and CB, FY2009FY2011
0 30% 31 40% 41 50% 51 60% 61 65% 66 70% 71 80% Total
MN1 MN2 MN3 MN4 MN5 MN6 MN7 MN8 MN9 MN10 MN11 MN12 Total
0 0 1 0 0 0 0 0 79 38 23 0 141
0 0 3 54 0 0 0 0 68 126 54 0 305
0 7 1,049 0 0 0 0 0 29 4 43 73 1,205
0 0 1,157 0 0 0 0 0 41 1 0 9 1,208
21
The use of the terms low , moderate , and middle income, is misleading and does not reect the actual income levels of New Yorkers.
This lack of low income housing in our wealthiest neighborhoods illustrates a mismatch while the City o en builds housing targe ng households with incomes higher than what is average for a neighborhood, the City rarely builds housing geared towards incomes below what is typical for neighborhood residents. In fact, only 2.9% of low and moderate income housing (120% AMI and below) was built in middle and upper income neighborhoods (120% AMI and above). These trends are similar in the outer boroughs as well. In Brooklyn, 8 of 18 Community Districts did not see a single unit developed for households earning less than 30% AMI. In the Bronx, 3 of 12 districts saw no housing for this popula on. In Queens and Staten Island, not only did 9 of 14 districts (Queens) and 2 out of 3 districts (Staten Island) not see any units for households earning below 30% AMI, there were only 13 units of housing for this popula on built in total throughout these two boroughs. It is a signicant accomplishment that a majority of NHMP units were built for low in come residents. But, it is a concern that data from the last two years suggest that most of these units were likely for the top earners in this category. More concerning however, is the use of the terms low , moderate , and middle income, to label AMI level categories. This labeling is misleading and does not reect the actual income levels of New Yorkers.
FIGURE4:
REAL AFFORDABILITY |
23
Two thirds of our aordable housing is actually unaordable to most neighborhood residents.
The result of this mismatch between AMI and actual median incomes is that much of the New Housing Marketplace has excluded truly low and moderate income New Yorkers. Because of the broad aordability categories provided by HPD for FY2004 FY2008, a detailed analysis of the en re New Housing Marketplace is impossible. We only have detailed aordability data for FY2009 FY2011. However, the data shows that about a third of the units under the NHMP between FY2009 FY2011 were available to those making the NYC median income and above. Furthermore, if we look at the community level, this mismatch becomes even starker. Since only maximum incomes, not minimum incomes (which vary according to program, as well as across the life of the New Housing Marketplace), were provided for the developments in our dataset, we have extrapolated minimum incomes needed to qualify for an aordable development by using a consistent discount of 10% of AMI o of the maximums. Using this method, we calculate that out of the 38,670 units developed NHMP between FY2009 and FY2011, only 13,053 of them (34%) were aordable to households making the median income or less for the typical household in their neighborhood. Two thirds of our aordable housing is actually unaordable to most neighborhood residents. Figure 5 illustrates the share of NHMP units developed that were aordable to the average household in each community district across the City. It shows that in a large share of the city, the vast majority of NHMP units were not aordable to the local community. In almost 1/3 of the neighborhoods (13/41) that have seen 100 units or more of aordable housing developed from FY2009 FY2011, 80% or more of these units are unaordable to the average local household. This is true across the City. We see it in some of the poorest neighborhoods, like Brownsville and Mo Haven, but also in middle class neighborhoods in Eastern Queens and the North Bronx, and even in one the Citys wealthiest neighborhoods, Greenwich Village.
24 | REAL AFFORDABILITY
FIGURE5:
All NHMP Starts FY2009 FY2011. Data Source: NYC HPD, NYCHDC, & United States Census American Community Survey
This disparity is greatest in the Citys lowest income communi es. These are areas that have historically experienced disinvestment and blight and are home to many of the Citys lower earning residents. Because of this history of disinvestment they o en also tend to be the areas with the most vacant land and development opportuni es, and hence, where the most aordable housing has been built. Because of the extremely low household incomes in these neighborhoods rela ve to the area median income, the City subsidized housing developed is o en unaordable to the overwhelming majority of their residents.
REAL AFFORDABILITY |
25
In half of the Community Districts, the average household cannot aord most of the aordable housing being developed in their neighborhood.
For example, in NYCs poorest borough, the Bronx, most of the units built under the early years of the Mayors plan have been targeted to households making less than 80% of AMI (see Table 9). In fact, in every Community District, at least 80% of housing developed was either for households earning less than 60% or 61% 80% of AMI and in eight of the twelve community districts, over 90% of the units developed were targeted to these households. However, in every community district in the Bronx the actual median household income is less than 80% of AMI. And in community districts 1 through 6, where almost four hs (79%) of the aordable housing in the Bronx has been built, the median household income is less than 40% of AMI, meaning the average household in these community districts does not make enough money to qualify for the vast majority of aordable housing that has been constructed under the New Housing Marketplace in their neighborhood.
TABLE 9:
BX1 BX2 BX3 BX4 BX5 BX6 BX7 BX8 BX9 BX10 BX11 BX12 Total BX
$20,037 $20,037 $21,617 $26,738 $26,131 $21,617 $29,179 $52,301 $38,475 $50,893 $45,276 $45,686
28.3% 28.3% 30.8% 37.9% 36.3% 30.8% 41.9% 78.4% 54.8% 76.5% 65.3% 64.4%
2,428 2,014 1,455 805 1,428 963 865 393 306 254 173 334 11,418
1.9% 2.1% 0.7% 9.2% 11.1% 17.1% 6.6% 56.2% 34.0% 97.2% 73.4% 33.8% 11.9%
The current City policy is to reserve half of all new development units for people living in the neighborhood. However, as the analysis above illustrates, in half of the Community Districts, the average household cannot aord most of the aordable housing being developed in their neighborhood. Out of the 41 community districts that saw signicant development (100 units or more) from FY2009 FY2011, in half (21 out of 41), the majority of aordable housing was unaordable to the average household. Given the recogni on that local residents should have access to local developments, the City must ensure that the aordable housing opportuni es are actually aordable to people in the community.
26 | REAL AFFORDABILITY
HUDs median income limits are not at all reec ve of the actual median income levels in New York City.
Source: United States Census Bureau. American Community Survey 1 year data for 2010. All income and household size data are 1 year ACS data unless otherwise noted. HUD Median Family Incomes are prorated for the average NYC household size for each year. 9 HUD only provides median incomes by family size, not overall median incomes. As such, to determine the overall median for HUD, we took the 2 person and 3 person limits, and prorated for the average household size in New York City, which was 2.64 people in 2010.
8
REAL AFFORDABILITY |
27
FIGURE 6:
This discrepancy between HUD income limits and NYC actual incomes holds true going back to at least 2005. Table 10 captures the consistent incongruity between HUD AMI in comparison to the actual NYC median incomes. Over the 6 year period from 2005 2010, on average, the actual New York City household median was only 76.4% of the HUD AMI levels.10
TABLE 10: 2005
2010 AordabilityComparison
2005 2006 2007 2008 2009 $66,482 $50,003 75% 2010 $68,456 $48,743 71% 76.4% Overall Average
HUDArea $54,169 $61,386 $61,457 $66,636 Median Income Actual NYC Median Household Income $43,434 $46,480 $48,631 $51,116 % of AMI 80% 76% 79% 77%
28 | REAL AFFORDABILITY
10
This problem is only further exacerbated by the enormous dierences in incomes across boroughs and neighborhoods. Median incomes in New York City vary greatly by borough, and even more so by Community District. Table 11 illustrates the dierences in the median incomes by borough in 2010. The Bronx and Brooklyns median incomes are far lower than those of Staten Island and Manha an. The Bronxs median income is less than half of the median income in Staten Island. This has signicant impacts on the aordability of NHMP in each borough. Staten Islands median income is consistent with HUD AMI and Manha ans is close. But, in the Bronx and Brooklyn, where the median incomes are substan ally lower than AMI, aordable housing would need to be developed at about the 50% and 70% AMI levels, respec vely, to be aordable to the median households in each borough.
TABLE 11: Median Income and Share of AMI by
Borough, 2010
To understand the very real implica ons of using HUDs Area Median Income limits, we can look at a typical aordable housing project built by LIHTC, the federal program that helps nance most of the newly constructed units developed by the City. The programs guidelines require that units serve households earning less than 60% AMI, which in 2010 was $47,520 for a family of four. However, if the actual median income of the City were used, the same unit would be available to a household of four earning as li le as $37,679. Furthermore, in New York Citys poorest borough, the Bronx, 60% of the median income for a four person household is just $24,627 a li le more than half of the income limit being used in the City aordable housing programs. The typical Bronx household would have to make 1.5 mes its income in order to be able to aord the majority of the aordable housing built in the Bronx. The mismatch between HUDs Area Median Incomes and the actual median incomes at the City and borough levels are especially true when looking neighborhood by neighborhood. Figure 7 shows each Community Districts median income as a percentage of AMI. Two thirds, or 40 of the 59 Community Districts in the City, have median incomes at or below the Low Income limit of 80% of AMI. And every single Community District in New York City, even the wealthiest, has a median income less than the Middle Income limit of 165% of AMI.
The typical Bronx household would have to make 1.5 mes its income in order to be able to aord the majority of the aordable housing built in the Bronx.
REAL AFFORDABILITY |
29
FIGURE7:
The high cost of opera ng housing in New York, paired with the extremely low incomes of many of its residents, o en makes it nancially unfeasible to build housing with large numbers of deeply low income units.
30 | REAL AFFORDABILITY
like energy eciency that can reduce opera ng expenses, there is s ll a baseline level, below which a development simply is not nancially sustainable, even without carrying debt. O en mes, this baseline is well above what the average households in low and very low income neighborhoods can aord. As a result, it is not always nancially feasible to build housing at the depth of aordability needed by the community. Furthermore, these nancial structures are constrained by the current Federal, State, and City policies and programs that fund aordable housing development, most notably the LIHTC program. The housing built under the New Housing Marketplace is, in many ways, reec ve of these constraints. And given these limita ons the aordable housing community has done an admirable job of trying to reach the needs of our low income residents. They highlight the reality that there are very real nancial structures that make it dicult to build at the aordability needed by low income residents. However, a sizeable por on of our communi es that have long been priced out of market rate housing are now being priced out of the aordable housing market as well. The City needs to have a housing policy that aligns with the needs of New Yorkers. The solu on cannot simply be to leave these families behind without aordable housing opportuni es. Nor is it sucient to rely solely on the federal government to house the millions of New Yorks very low income residents. Instead, buildings in higher income neighborhoods have rents that can support the opera ng costs of lower income housing units. And further thinking is needed on how income mixing in buildings can be be er used to cross subsidize larger numbers of units aordable to the one third of New Yorkers who make less than 40% of AMI a level that is widely considered too low to support a building on its own.
A sizeable por on of our communi es that have long been priced out of market rate housing are now being priced out of the aordable housing market as well.
REAL AFFORDABILITY |
31
32 | REAL AFFORDABILITY
Moving forward, the City has an opportunity to recommit to aordable housing actually being aordable for local residents. The con nued demand by developers for City subsidies and land needs to be leveraged into deeper aordability for residents. The City has demonstrated that when it sets a housing policy goal, it can meet it. This has been true of unit targets under New Housing Marketplace, and it can be true of aordability targets under the Citys next mayor. The next administra on has the ability to build not just housing, but truly aordable housing that stabilizes our communi es.
REAL AFFORDABILITY | 33
Star ng in 2017, NYC will begin losing an average of more than 11,000 affordable units a year.
In the late 1980s, two things happened to catalyze aordable housing development the Low Income Housing Tax Credit Program, and Mayor Kochs Ten Year Housing plan. During the early years of these, aordability restric ons were only for 15 years. We soon learned this wasnt sucient, and the LIHTC requirements were extended to 30 years. Under Mayor Bloomberg, the City has gradually moved toward a policy of incen vizing 50 or 60 years of aordability, although in recent years they have dialed that back and only been requiring the 30 year short term aordability. Clearly, long term and, especially, permanent aordability is a much be er use of public subsidy than units only aordable in the short term.
34 | REAL AFFORDABILITY
In addi on, long term and permanent aordability does not cost the City addi onal funds; it can be mandated as part of the aordable housing program. For example, in 2011 when the City incen vized 60 years of aordability for its main development program, Low Income Housing Tax Credits, 100% of the developments pledged 60 years of aordability. The City managed to double its aordable housing value with a stroke of the pen, all without any addi onal public subsidy. A comprehensive analysis of aordability lengths for City sponsored aordable housing developments from 1987 2007 is available in ANHDs 2010 report, A Permanent Problem Needs A Permanent Solu on: New York Citys Next Aordable Housing Expiring Use Crisis and the Need for Permanent Aordability.
35
Comprehensive community development has been le to mission oriented organiza ons, o en not for prot developers, whose broader contributions to the neighborhood need to be recognized and encouraged.
36 | REAL AFFORDABILITY
These contribu ons also need to be recorded and incorporated as does crea ng local retail, community, and educa onal space, bringing services such as a er school care and voca onal training to neighborhoods, and crea ng local jobs and suppor ng local businesses. The City needs to approach development comprehensively, considering and lling other community needs as well as the need for aordable housing.
REAL AFFORDABILITY |
37
38 | REAL AFFORDABILITY
Looking Ahead:
The next mayor of New York City has an extraordinary opportunity the rst in over a decade to reshape the future of aordable housing development policy in our city to strengthen the impact on families and communi es. It is impera ve that he or she be er align aordable housing programs and policies to reect the needs of our residents and communi es. Neighborhoods have dierent needs. Some desperately need community services and retail op ons along with aordable housing. Others need more housing for large families or single senior ci zens. There are neighborhoods that are rapidly becoming more expensive and need their aordable housing to stay aordable for future gener a ons as well as the current residents. And yet others where the residents need deeply aordable housing for working families. It is not enough to simply produce units. Our aordable housing must be as unique as the communi es and residents it serves. A one size ts all solu on doesnt t New Yorkers, and it wont t as an aordable housing solu on. Addressing the Citys growing aordable housing crisis must be a comprehensive, nuanced, and three dimensional approach for the future. ANHD is asking the ques on: What if we set dierent goals? What if we set the goal of building the kind of housing and neighborhoods that communi es need? Units Produced is an easy measurement to quan fy quan fying Real Aordability is more dicult. But s ll, a large reason behind the success of the New Housing Marketplace was that it set a goal, and con nually measured progress toward that goal. We need a new goal. The City needs a new way of measuring success and measuring a return on public investment that takes into account the quality, not just quan ty, of aordable housing built.
The City needs a new way of measuring success and measuring a return on public investment that takes into account the quality, not just quan ty, of aordable housing built.
REAL AFFORDABILITY |
39
40 | REAL AFFORDABILITY
As the box above illustrates, the Real Aordability Index is calculated by mul plying the number of people housed in a unit by the number of years a unit is required to be aordable, divided by the percent of Area Median Income that the unit serves. More specically, these inputs are: Number of Units is simply the number of units in the development project. Number of People Served is calculated using 1.5 people per bedroom, and 1 for a studio, unless the size of individual households is available. Years of Aordability is based on the signed regulatory agreement, unless the primary sponsor has a mission driven commitment to provide aordable housing for low income people as part of its Ar cles of Incorpora on, in which case aordability is assumed to be permanent. ANHD is using 99 years in the equa on in lieu of Permanent Aordability. Average % of AMI is calculated by using the average income restric ons as they relate to the units. It should be noted that in many developments, most notably in tenant preserva on and suppor ve housing programs, the actual income of some tenants is o en lower than the income restric ons on the unit. If there is a lower income household in the unit for which informa on is available, we use the lower income, but only if the highest registered legal rent of the unit also reects this income. We do not count rent or opera on subsidies in our calcula on of income restric ons or public subsidy invested. It should also be noted that a er ini al qualica on, tenants may generally earn any amount of money in subsequent years while s ll living in an income restricted unit. Community Impact reects the concrete value a development brings to its residents and surrounding community beyond simply housing. Many aordable developments create much needed retail op ons, social services, economic development or community facili es in addi on to housing units. Because this impact is not a per unit calcula on, instead being an overall value brought to the community, it is quan ed through a 10 point booster assigned to the development as a whole. Below are two theore cal examples of how the Real Aordability Index can be used to calculate the amount of public benet secured by the City. Example 1 is an aordable housing development that has 10 studio units serving singles, each earning 60% AMI, with a 30 year regulatory agreement, and no other components. While it has 10 units, the Real Aordability impact for this development is only 5. Example 2 is a development that has 10 two bedroom units serving households of 3 each earning 40% AMI, with a 60 year regulatory agreement. In addi on, the development has a much needed child care facility on the ground oor. Here, even though it has the same 10 units, the Real Aordability Index score is 55. The larger score demonstrates the be er return to the City based on serving three mes as many people, at a deeper aordability level, for twice as long of an aordability period, with an addi onal community component.
REAL AFFORDABILITY |
41
EXAMPLE1:
10 (____) 100
EXAMPLE2:
10 (____) 100
While Example 2 does have a much higher Real Aordability Index score, this does not necessarily mean that the second project is a be er public investment that the rst. The second project almost certainly requires more public subsidy, as larger and more deeply aordable units all take more subsidy per unit. Longer term aordability can some mes require slightly greater public investment, and the addi onal retail component takes away square footage from the building, reducing its poten al residen al units. If the second development costs more than 11 mes the public subsidy of the rst development, the rst development is a be er public investment.
42 | REAL AFFORDABILITY
Below we calculate the tradi onal number of units built per dollar of public subsidy analysis and compare this to the Return on Real Aordability. We compare two sepa rate 8 unit developments, one at 1264 and 1279 Decatur Street in Brooklyn, developed under the Par cipa on Loan Program; the other at 459 West 147th Street in Manha an, developed under the Small Building Loan Program. From this we can determine how ecient the project is in terms of the public benet provided for the public subsidy spent. Tables 12 and 13 provide a snapshot of the Decatur Street buildings. As Table 12 details, there are 8 units housing 30 people. Half are two bedroom, and half are three bedroom apartments. In terms of aordability, the units have a wide range, with a quarter at 50% AMI, half at 60% AMI, and a quarter at 120% AMI. The Decatur Street project does not have a community component.
TABLE 12: Unit Size by # of Bedrooms TABLE 13:
Decatur Street
Number of Units
Decatur Street 0 BR 1 BR 2 BR
30%
3 BR
4 4 8 8
Tables 14 and 15 below provide a snapshot of 459 West 147th street. As Table 14 details, there are 8 units, all studios, housing 8 people. In terms of aordability, three of the studios are at 60% AMI, and the other 5 are at 80% AMI. The West 147th Street project also does not have a community component.
TABLE 14: Unit SIze by # of Bedrooms
TABLE 15: Targeted AMI by Unit SIze
Number of Units 8
1 BR
2 BR
3 BR
8 8
REAL AFFORDABILITY |
43
A side by side comparison of the Citys return on public investment for these two developments is captured below in Table 16. The Decatur Street project was developed with $151,411 in City Capital, and $1,050,000 in HOME funds, for a total of $1,201,411 in public subsidy, or $150,176 per unit. 459 West 147th Street was developed with $299,915 in City Capital and $302,500 in HOME funds, for a total of $602,415 in public subsidy, or $75,302 per unit. By the New Housing Marketplaces index, units, the West 147th Street building is twice as good of an investment as the Decatur Street development.
TABLE 16:
Number of Units City Capital HOME Funds Total Subsidy Subsidy per Unit Real Aordability Score Subsidy per Unit of Real Aordability (Return on Real Aordability)
However, when measuring return on Real Aordability, a dierent story becomes clear. The large units, combined with the deeper aordability, gives the Decatur Street a Real Aordability index of 12.4. In contrast, the studio apartments of the West 147th street development, as well as the slightly higher aordability levels, give it a score of 4.3. While each unit produced in the Decatur Street building cost the City $150,176, each units Real Aordability cost to the City is only $81,314. At West 147th Street, while each unit cost the City $75,302, each units Real Aordability cost to the City is $178,493 over twice that of the Decatur Street building. So while at rst glance the West 147th street development seems like twice as ecient an investment as the Decatur Street one, with a closer look, we can see the opposite is actually true.
44 | REAL AFFORDABILITY
Closing
ANHDs Real Aordability Index will allow community leaders to advocate for
the produc on of the types of units most needed in their communi es. For instance, a community that needs larger family units would be able to point out that a develop ment that consisted of a smaller number of three bedroom units has a higher Real Aordability Index score than a development of a greater number of studio apartments, or a community that needs very low income housing would be able to point out that a smaller number of very low income units has a higher Real Aordability Index score than a greater number of middle income units. This lets advocates, developers, and City ocials shape the best housing policy for the City and its communi es, instead of solely focusing on maximizing the units produced. Indeed, it is precisely these types of conversa ons that ANHD hopes to facilitate with candidates and the community during the next mayoral elec on.
REAL AFFORDABILITY |
45
Appendices
Methodology New Housing Marketplace Data
ANHD obtained data on geography, unit size, and depth of aordability of the New Housing Marketplace through a Freedom of Informa on Act (FOIA) request from the Department of Housing Preserva on and Development (HPD) and the Housing Devel opment Corpora on (HDC). Because of the way HPD and HDC track the details of their units produced, dierent informa on was available for dierent components of the New Housing Marketplace. This leads to dierent datasets being used for dierent areas of analysis. The largest dataset used is the en re 124,418 units of aordable housing developed from FY2004FY2011 under the New Housing Marketplace, for which ANHD received informa on on geography and broad aordability levels (<60% AMI, 60% 80% AMI, 80% 120% AMI, and 120% 180% AMI). Analysis regarding geography of the New Housing Marketplace and broad aordability levels uses this dataset, except where indicated in the report. This dataset is shown in full as two appendices, one by geography and one by aordability. A subset of this overall dataset 38,670 units, which is the total developed under the New Housing Marketplace from FY2009FY2011 includes narrower aordability levels. These consist of <30% AMI units, and then 10% AMI increments up to the 180% AMI level, as well as a 60 65% AMI level, and 65% 70% AMI level. There are also three other categories provided: Non Restricted, Other, and Super Units (usually rent free units reserved for the superintendent of the building). When analyzing aordability levels, ANHD eliminated Super Units and Other from the dataset, and counted Non Restricted units as being at the highest aordability level (180% AMI). This dataset is shown in full as an appendix. Through the course of the New Housing Marketplace, HPD has not consistently tracked unit size. A dataset of 53,211 units was provided that contained specic unit sizes, which was cross referenced with geography, but not with aordability. The sample includes the following units: all HPD Preserva on and HPD New Construc on projects started prior to July 2010, with a group of pre July 2010 New Construc on projects not included if they had completed in the last two years. ANHD paid HPD to re congure its database to obtain some of this data. This dataset is shown in full as an appendix. Since our data request and recongura on, HPD has improved its database rela ng to unit size, and more comprehensive data may now be available.
46 | REAL AFFORDABILITY
A small dataset of 9,970 units, was also provided which cross references unit size with the narrow AMI increments. This data is from a web based HPD tracking applica on called OnTrack used by HPDs Oce of Development to track the loan closing process. The dataset is for New Construc on projects started a er July 2010, and a group of pre July 2010 New Construc on starts if they completed in the past 2 years. This dataset was not included in our analysis of unit size in the NHMP, as it was limited to New Construc on only, but was presented in full and analyzed separately in the Housing for What Size Family? sec on. For length of aordability, ANHD referred to our previous analysis, rst published in the spring of 2010 in our report, A Permanent Problem Requires a Permanent Solu on: New York Citys Next Aordable Housing Expiring Use Crisis and the Need for Permanent Aordability, which tracks the length of aordability for city sponsored developments from 1987 2007. Finally, ANHD led a separate Freedom of Informa on Request for specic HPD and HDC projects chosen at random, for which we received detailed informa on on the specic projects, including unit sizes, depth of aordability, and amount and type of public subsidy invested. Two of these projects are used as examples to calculate our Return on Real Aordability, as a new way of examining the specic impact of each public dollar spent on aordable housing. We also received detailed informa on on public subsidy invested for all HDC projects.
Demographic Data
All demographic and income informa on comes from the United States Census Bureaus American Community Survey (ACS). This is sample data, and has a margin of error associated with it which is not displayed in our analysis. For median income levels and household sizes at the County (Borough) and City level, we used 1 year ACS data. ACS data is from calendar year 2010 unless otherwise noted. For geographies smaller than the County (Borough) level, 1 year ACS data is not available. For median income and household size at the Community District level, we used 5 year ACS data from 2006 2010, which is a sample of households taken over all ve years, with income informa on adjusted for ina on to 2010 dollars. For 5 year data, the US Census uses Public Use Microdata Areas (PUMAs), which are approximately analogous to New York Citys Community Districts, although not completely coterminous. For purposes of this analysis, we treated PUMAs and Community Districts as analogous.
REAL AFFORDABILITY |
47
Less than $10,000 $10,000 to $14,999 $15,000 to $24,999 $25,000 to $34,999 $35,000 to $49,999 $50,000 to $74,999 $75,000 to $99,999 $100,000 to $149,999 $150,000 to $199,999 $200,000 or more
We then determined the corresponding income for each 10% AMI band (prorated for the average household size), assumed incomes were evenly distributed throughout each band provided by the ACS, and re allocated them accordingly. We express this as the top end of the 10% income band, so the percentage of New Yorkers at the 60% AMI level is the percentage of New Yorkers making between 50% and 60% of AMI.
48 | REAL AFFORDABILITY
Determining Aordability
When determining depth of aordability, HPD and HDC provided upper income limits for the various units, expressed as a percentage of Area Median Income. For instance, a 60% AMI unit means that nobody making over 60% of the Area Median Income is allowed to rent the apartment. However this does not indicate the lowest possible income a household could have and s ll qualify to rent the unit. While the upper income limit for units developed under the New Housing Marketplace has consistently been tracked as a percentage of AMI, there has not been a consistent methodology for determining the lower income limit of these income bands. Depending on program, year, and level of aordability, dierent methodologies have been used, resul ng in many dierent income band sizes. In this report we have applied a standard 10% AMI income band to all NHMP units i.e., we count a 60% AMI unit as being aordable to anyone making between 50% and 60% of AMI. This also corresponds to the 10% cohorts used to analyze the New Housing Marketplace, as well as the incomes of New York City households, as demonstrated above.
REAL AFFORDABILITY |
49
ANHD analyzes the aordability of the New Housing Marketplace by unit, not by household. In calcula ng aordability, we refer exclusively to the units income restric ons provided to us by HPD and HDC. It should be noted that in many developments, most notably suppor ve housing developments, the actual income of some tenants is o en lower than is indicated by the income restric ons on the unit, usually due to rental or other opera onal subsidies making up the rent dieren als. In other preserva on programs where tenants are already in place, incomes and rents may be lower for the current household occupying the unit than are reected in the overall restric ons on the unit. Because of this, it is important to note that the lack of very low income apartments developed under the New Housing Marketplace does not necessarily reect a lack of very low income New Yorkers currently served by these developments. Conversely, it should also be noted that a er ini al qualica on tenants may generally earn any amount of money in subsequent years while still living in an income restricted unit.
50 | REAL AFFORDABILITY
New Housing Marketplace Units by Area Median Income (AMI) FY2004 FY2011
PRE FY2009 Percent AMI 0 60% 61 80% 81 100% 101 120% 121 180% Non Restricted Super Total Pre FY2009 All NHMP 0 60% 61 80% 81 100% 101 120% 121 180% Non Restricted Super Other Total NHMP Number 42,513 60,580 6,356 1,842 10,431 2,273 182 241 124,418 % of NHMP 34.2% 48.7% 5.1% 1.5% 8.4% 1.8% 0.1% 0.2% 100.0% Number 19,951 53,980 4,944 1,043 5,830 0 0 85,748 % of NHMP 23.3% 63.0% 5.8% 1.2% 6.8% 0.0% 0.0% 100.0% POST FY2009 Number 0 30% 31 40% 41 50% 51 60% 61 70% 71 80% 81 90% 91 100% 101 110% 111 120% 121 130% 131 141 151 161 140% 150% 160% 170% 1,389 1,660 4,606 14,907 2,895 3,705 437 975 341 458 773 2,457 91 86 704 490 0 2,273 182 241 38,670 % of NHMP 3.6% 4.3% 11.9% 38.5% 7.5% 9.6% 1.1% 2.5% 0.9% 1.2% 2.0% 6.4% 0.2% 0.2% 1.8% 1.3% 0.0% 5.9% 0.5% 0.6% 100.0%
171 180% 181 250% Non Restricted Super Other Total Post FY2009
REAL AFFORDABILITY |
51
UNITS
2,891 2,933 3,174 2,330 5,269 363 451 1,468 1,406 302 21 252 3,109 459 1,226 4,179 233 158 30,224
% BORO
9.6% 9.7% 10.5% 7.7% 17.4% 1.2% 1.5% 4.9% 4.7% 1.0% 0.1% 0.8% 10.3% 1.5% 4.1% 13.8% 0.8% 0.5% 100.0%
% CITY
2.3% 2.4% 2.6% 1.9% 4.2% 0.3% 0.4% 1.2% 1.1% 0.2% 0.0% 0.2% 2.5% 0.4% 1.0% 3.4% 0.2% 0.1% 24.3%
QUEENS
COMMUNITY DISTRICT
QN1 QN2 QN3 QN4 QN5 QN6 QN7 QN8 QN9 QN10 QN11 QN12 QN13 QN14 TOTAL Astoria / Long Island City Sunnyside / Woodside E. Elmhurst / Jackson Hts. Corona / Elmhurst / Lefrak City Maspeth / Mid Village / Ridgewood Forest Hills / Rego Pk Flushing / Bay Terrace /Whitestone Jamaica / Fresh Meadows Kew Gardens / Ozone Pk Howard Beach / Ozone Pk Douglaston / Li le Neck Hollis / Rochdale / St. Albans Bellerose / Laurelton / Queens Vil Arverne / Far Rockaway / Seaside
UNITS
519 1,092 59 37 28 76 453 2,598 165 42 23 1,361 98 5,392 11,943
% BORO
4.3% 9.1% 0.5% 0.3% 0.2% 0.6% 3.8% 21.8% 1.4% 0.4% 0.2% 11.4% 0.8% 45.1% 100.0%
% CITY
0.4% 0.9% 0.0% 0.0% 0.0% 0.1% 0.4% 2.1% 0.1% 0.0% 0.0% 1.1% 0.1% 4.3% 9.6%
52 | REAL AFFORDABILITY
BRONX
COMMUNITY DISTRICT
BX1 BX2 BX3 BX4 BX5 BX6 BX7 BX8 BX9 BX10 BX11 BX12 TOTAL Melrose / Mo Haven Hunts Pt / Longwood Crotona Pk E / Morrisania Concourse / Highbridge / Mt. Eden Fordham / Mt. Hope / Univers Hts Belmont / East Tremont Bedford Pk / Kingsbr Hts / Norwood Fieldston / Riverdale Castle Hill / Parkchester Soundview Co Op City / Pelham Bay / Throgs Neck Morris Park / Pelham Parkway Baychester / Williamsbridge
UNITS
6,545 4,245 6,005 3,181 4,098 3,567 3,430 563 3,377 949 991 892 37,843
% BORO
17.3% 11.2% 15.9% 8.4% 10.8% 9.4% 9.1% 1.5% 8.9% 2.5% 2.6% 2.4% 100.0%
% CITY
5.3% 3.4% 4.8% 2.6% 3.3% 2.9% 2.8% 0.5% 2.7% 0.8% 0.8% 0.7% 30.4%
MANHATTAN
COMMUNITY DISTRICT
MN1 MN2 MN3 MN4 MN5 MN6 MN7 MN8 MN9 MN10 MN11 MN12 TOTAL FiDi / Tribeca W. Village / Soho Chinatown / LES Chelsea / Clinton Union Sq / Midtown East Stuy Town / Turtle Bay UWS / Manha an Valley UES / Yorkville Hamiliton Hts / W. Harlem Central Harlem East Harlem Inwood / Washington Hts.
UNITS
28 1,391 7,881 6,869 99 637 2,727 511 2,598 10,352 7,829 1,540 42,462
% BORO
0.1% 3.3% 18.6% 16.2% 0.2% 1.5% 6.4% 1.2% 6.1% 24.4% 18.4% 3.6% 100.0%
% CITY
0.0% 1.1% 6.3% 5.5% 0.1% 0.5% 2.2% 0.4% 2.1% 8.3% 6.3% 1.2% 34.1%
STATENISLAND
COMMUNITY DISTRICT
SI1 SI2 SI3 TOTAL TOTAL CITYWIDE Stapleton / St. George South Beacon / Willowbrook To enville / Great Kills
UNITS
1,774 137 35 1,946 124,418
% BORO
91.2% 7.0% 1.8% 100.0% 100%
% CITY
1.4% 0.1% 0.0% 1.6% 100%
REAL AFFORDABILITY |
53
New Housing Marketplace Units by Size and Community District, FY2004 FY2010
BROOKLYN
CB BK1 BK2 BK3 BK4 BK5 BK6 BK7 BK8 BK9 BK10 BK11 BK12 BK13 BK14 BK15 BK16 BK17 BK18 TOTAL Percent 0 BR 226 644 210 103 468 3 52 39 30 0 0 12 57 105 0 350 46 0 2,345 16.1% 1 BR 419 526 534 559 373 20 95 129 197 0 0 0 443 58 0 746 15 1 4,115 28.3% 2 BR 551 243 630 641 605 15 36 271 58 0 0 1 778 20 0 954 75 3 4,881 33.6% 3 BR 247 216 267 207 341 52 5 99 79 0 0 1 0 29 0 540 34 2 2,119 14.6% 4 BR 24 43 19 11 69 2 0 3 0 0 0 0 0 12 0 68 2 0 253 1.7% 5 BR 0 1 2 7 12 0 0 0 0 0 0 0 531 0 0 2 0 0 555 3.8% 6 BR 0 0 1 3 3 0 0 0 0 0 0 0 0 0 0 0 0 0 7 0.0% Unknown 0 21 160 49 1 0 0 0 1 0 0 0 7 0 0 9 0 0 248 1.7% TOTAL 1,467 1,694 1,823 1,580 1,872 92 188 541 365 0 0 14 1,816 224 0 2,669 172 6 14,523 100.0%
QUEENS
CB QN1 QN2 QN3 QN4 QN5 QN6 QN7 QN8 QN9 QN10 QN11 QN12 QN13 QN14 TOTAL Percent 0 BR 32 0 0 0 0 0 0 0 0 0 0 41 0 4 77 0.0% 1 BR 230 16 3 1 0 0 0 81 12 1 0 105 2 171 622 3.2% 2 BR 39 0 9 8 2 0 4 0 1 4 0 78 5 696 846 25.6% 3 BR 2 1 0 1 1 0 6 0 2 8 0 87 7 762 877 34.8% 4 BR 0 0 0 0 0 0 0 0 0 1 0 8 0 2 11 36.0% 5 BR 0 0 0 0 0 0 0 0 2 0 0 0 0 0 2 0.5% 6 BR 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.1% Unknown 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.0% TOTAL 303 17 12 10 3 0 10 81 17 14 0 319 14 1,635 2,435 100.0%
54 | REAL AFFORDABILITY
BRONX
CB BX1 BX2 BX3 BX4 BX5 BX6 BX7 BX8 BX9 BX10 BX11 BX12 TOTAL Percent 0 BR 561 170 433 234 124 338 290 84 2 35 74 12 2,357 13.4% 1 BR 797 643 686 792 594 450 606 39 298 353 45 333 5,636 32.0% 2 BR 1,305 987 1,070 711 677 584 226 24 117 232 45 151 6,129 34.8% 3 BR 513 355 224 220 151 56 58 5 247 70 7 43 1,949 11.1% 4 BR 126 91 37 14 20 1 1 0 116 0 0 0 406 2.3% 5 BR 10 14 0 0 0 0 0 0 0 0 0 0 24 0.1% 6 BR 2 6 0 0 0 0 0 0 0 0 0 0 8 0.0% Unknown 3 170 7 252 169 526 0 0 0 0 0 0 1,127 6.4% TOTAL 3,317 2,436 2,457 2,223 1,735 1,955 1,181 152 780 690 171 539 17,636 100.0%
MANHATTAN
CB MN1 MN2 MN3 MN4 MN5 MN6 MN7 MN8 MN9 MN10 MN11 MN12 TOTAL Percent 0 BR 4 15 739 620 0 36 379* 76 259 1,156 354 223 3,861 21.4% 1 BR 24 5 1,517 1,020 0 174 151 156 283 1,486 838 262 5,916 32.7% 2 BR 0 1 1,068 396 0 15 76 21 558 2,346 820 311 5,612 31.0% 3 BR 0 0 328 2 0 0 30 1 373 868 290 164 2,056 11.4% 4 BR 0 0 0 0 0 0 0 0 132 217 35 10 394 2.2% 5 BR 0 0 0 0 0 0 0 0 17 34 2 10 63 0.3% 6 BR 0 0 0 0 0 0 0 0 2 1 0 0 3 0.0% Unknown 0 0 0 0 0 0 0 0 60 103 3 8 174 1.0% TOTAL 28 21 3,652 2,038 0 225 636 254 1,684 6,211 2,342 988 18,079 100.0%
STATENISLAND
CB SI1 SI2 SI3 TOTAL Percent 0 BR 72 29 0 101 18.8% 1 BR 138 63 0 201 37.4% 2 BR 126 12 0 138 25.7% 3 BR 87 0 2 89 16.5% 4 BR 8 0 0 8 1.5% 5 BR 0 0 0 0 0.0% 6 BR 0 0 0 0 0.0% Unknown 0 1 0 1 0.2% TOTAL 431 105 2 538 100.0%
CITYWIDE
0 BR TOTAL Percent 8,741 16.4% 1 BR 16,490 31.0% 2 BR 17,606 33.1% 3 BR 7,090 13.3% 4 BR 1,072 2.0% 5 BR 644 1.2% 6 BR 18 0.0% Unknown 1,550 2.9% TOTAL 53,211 100.0%
REAL AFFORDABILITY |
55
56 | REAL AFFORDABILITY
REAL AFFORDABILITY |
57
NONPROFIT ORG
50 Broad Street, Suite 1125 New York, NY 10004 2376 www.anhd.org p: 212.747.1117 f: 212.747.1114 e: info@anhd.org
US POSTAGE PAID
NEW YORK, NY PERMIT No. 3933